Token-Efficient Memory: How Aura Caches BM25, Repo Maps, and Long-Term Context
Three layers of on device persistence that cut token waste and keep the AI focused between sessions.
May 12, 20266 min read12
Command Palette
Search for a command to run...
Series
Aura IDE Development Series
Architecture decisions, implementation deep-dives, and release notes for Aura — a local-first AI coding IDE with a Planner/Worker agent split, MCP integration, and persistent project memory.
Building Aura: An Open-Source AI Coding Harness You Can Control
Own the Harness: An Open Source AI Coding Tool Where You See the Plan, Inspect the Edits, and Trust the Result
Jun 4, 20268 min read
How I Used Aura to Build a Lead Discovery Tool Without Burning Tokens
Signal Desk finds relevant conversations, ranks them as leads, writes reviewable inbox cards, and helps draft outreach replies. I built it with Aura’s phased workflow, visible diffs, validation, and cheap models for about ten cents.
Jun 7, 20269 min read6
From Chat Box to Orchestration: How Aura Became a Real AI Coding IDE
How Drones and a mobile companion turned Aura into a local AI orchestration IDE.
Jun 9, 20267 min read7
I Built a Coding Harness, Then It Started Improving Itself
A couple months ago, Aura was basically a chat box with ambition. One window. One model. A pile of ideas. A lot of duct tape. It could help, sure, but it still felt like every other AI coding setup I
Jun 22, 20268 min read1